For over half a century now, primate vocalizations have been thought to undergo little or no experience-dependent acoustic changes during development. If any changes are apparent, then they are routinely (and quite reasonably) attributed to the passive consequences of growth. Indeed, previous experiments on squirrel monkeys and macaque monkeys showed that social isolation, deafness, cross-fostering and parental absence have little or no effect on vocal development. Here, we explicitly test in marmoset monkeys—a very vocal and cooperatively breeding species—whether the transformation of immature into mature contact calls by infants is influenced by contingent parental vocal feedback. Using a closed-loop design, we experimentally provided more versus less contingent vocal feedback to twin infant marmoset monkeys over their first 2 months of life, the interval during which their contact calls transform from noisy, immature calls to tonal adult-like “phee” calls. Infants who received more contingent feedback had a faster rate of vocal development, producing mature-sounding contact calls earlier than the other twin. The differential rate of vocal development was not linked to genetics, perinatal experience, or body growth; nor did the amount of contingency influence the overall rate of spontaneous vocal production. Thus, we provide the first experimental evidence for production-related vocal learning during the development of a nonhuman primate.